Field Notes Journal

Field Notes Journal Entry

Towards a Seasonal Typology

Entry dated 5 April 2026 · Author: David Walker

Grouping familiar species not by name, but by the shape of their year

Category: wildlife

From Individual Patterns to Types

In recent entries in this series, I’ve been looking at individual species — blackbird, robin, chiffchaff — and how their patterns emerge when observations are grouped by month.

Those pieces considered each species on its own terms, its own rhythm across the year, but, having worked through a number of them, a further question suggested itself: If these patterns can be seen for individual species, do they begin to fall into recognisable types when viewed together?

A Simple Approach

The approach taken here is deliberately simple: For each species, observations are grouped by month, producing two basic measures: total sightings, and the number of days on which the species was recorded. From these, a small number of derived features can be calculated — how many months the species appears in, how long it is absent, where its activity is centred in the year, and how variable its presence is.

From that, a classification can be attempted. Not a statistical model, and not something trained or optimised, but a set of simple, interpretable rules. The aim is not to predict, but to describe — to see whether the shapes already visible in the data can be given names.

What emerges is a small typology of seasonal behaviour, as the following table for a selection of locally encountered species demonstrates:

Species Classification
Blackbird (Turdus merula) Detectability-driven resident
Blackcap (Sylvia atricapilla) Partial migrant
Chiffchaff (Phylloscopus collybita) Summer visitor
Common Starling (Sturnus vulgaris) Aggregation-driven resident
Fieldfare (Turdus pilaris) Winter visitor
Mistle Thrush (Turdus viscivorus) Broad seasonal curve
Redwing (Turdus iliacus) Winter visitor
Robin (Erithacus rubecula) Detectability-driven resident
Swallow (Hirundo rustica) Summer visitor
Woodpigeon (Columba palumbus) Aggregation-driven resident
Wren (Troglodytes troglodytes) Detectability-driven resident

There are species that are present throughout the year, but vary in how readily they are encountered — blackbird, robin, wren — which can be thought of as detectability-driven residents.

Others are present year-round but fluctuate more in number than in occurrence — starling, woodpigeon — reflecting patterns of gathering and dispersal. These might be described as aggregation-driven residents.

Then there are the more familiar seasonal patterns: summer visitors such as chiffchaff, swallow and swift, and winter visitors such as fieldfare and redwing, where the period of absence is clearly visible.

And, perhaps most interestingly, there are species that sit between these categories — most notably blackcap — where a strong summer presence is accompanied by a smaller but persistent winter signal. These begin to suggest a class of partial migrants.

Patterns Become Categories

What is striking is not the complexity of the method — there is very little of that — but how readily these categories emerge from simple summaries.

The shapes were already present in the earlier charts. The classification is, in effect, just a way of naming them but, once named, they allow the data to be read not just as a record of what’s present but as a set of patterns indicative of how each species moves through the year:

Classification Pattern
Aggregation-driven resident Present through most of the year, with counts varying more strongly than presence
Broad seasonal curve Activity is spread across a broad part of the year
Detectability-driven resident Present through most of the year, but encounter frequency varies seasonally
Partial migrant Summer-centred pattern with a persistent winter tail
Resident Present through most of the year with relatively even seasonal behaviour
Sparse / low-signal Insufficient seasonal signal for confident classification
Summer visitor Strong seasonal absence with activity centred in spring/summer
Winter visitor Strong seasonal absence with activity centred in winter

Limitations and Observational Bias

There are, however, clear limitations.

The classifications are only as reliable as the observations that underpin them.

Some species are recorded frequently and consistently, and their patterns are well defined. Others appear only occasionally, or unevenly, and for these the data is too sparse to support any confident interpretation. These have been set aside as low signal — not absent, but not yet sufficiently familiar.

A Tawny Owl, for instance, appears in these records as a winter visitor. Not because it departs in summer, but because it is relatively infrequently encountered in these records and the sightings that are present have occurred most often during the colder months.

The classification, in that case, is not wrong — but it is describing something slightly different from biological presence. It is describing the intersection between species behaviour and human observation.

Patterns of Encounter

With that in mind, the patterns being uncovered here are not purely ecological, nor purely observational, but a combination of the two. They are, in a sense, records of encounter.

That idea returns to something noted in the previous piece: that these are not abstract descriptions of species, but patterns derived from repeated encounters with them in a particular place.

The classifier does not know what a blackbird or a swallow is. It only knows how often, and when, they have been seen yet, from that, something recognisable emerges.

What This Does (and Doesn’t) Do

It would be easy to overstate what has been achieved here. The method is simple, the rules are heuristic, and the results are dependent on the completeness and consistency of the underlying observations. But within those limits, it appears to be doing something useful.

It provides a way of stepping back from individual records, and seeing instead the shape of the year as experienced through familiar species. Not as a list, but as a set of recurring patterns.

Looking Ahead

It will be interesting to see how far this approach can be taken.

Whether the categories hold as more species are added, whether they need refining, and where they begin to break down.

For now, though, it offers a different way of looking at the data: Not just what was seen, and when, but the form those sightings take, when viewed together.